Lilly Tech Systems
AI Supply Chain Security
Secure every link in the AI pipeline. Learn to protect model provenance, verify data lineage, manage dependency risks, detect poisoned models, and implement SBOM for machine learning systems.
Your Learning Path
Follow these lessons in order, or jump to any topic that interests you.
1. Introduction
What is AI supply chain security? Understanding the attack surface across the ML pipeline and why it matters.
2. Model Supply Chain
Model provenance, pre-trained model risks, poisoned models on Hugging Face, model signing, and verification.
3. Data Supply Chain
Data lineage tracking, dataset poisoning, data provenance, integrity verification, and trusted data sources.
4. Dependency Risks
ML library vulnerabilities, typosquatting, dependency confusion, SBOM for ML, and secure package management.
5. Verification
Cryptographic model signing, hash verification, reproducible builds, attestation frameworks, and trust chains.
6. Best Practices
End-to-end supply chain hardening, governance frameworks, continuous monitoring, and compliance strategies.
What You'll Learn
By the end of this course, you'll be able to:
Trace Model Provenance
Track the origin, training data, and modification history of ML models throughout their lifecycle.
Secure Data Pipelines
Implement data lineage tracking and integrity verification to prevent data poisoning attacks.
Manage Dependencies
Create and maintain Software Bill of Materials for ML systems and mitigate dependency risks.
Verify Model Integrity
Apply cryptographic signing, hash verification, and attestation to ensure model authenticity.